Better prompt caching for GPT-6
This is mainly relevant if your organisation builds tools or integrations on OpenAI's API — it lowers the cost of repeated or long-running AI agent calls by reusing cached context, with a new dashboard to track savings.
What changed
Source summary
OpenAI improved prompt caching for the GPT-6 model family, increasing cache hit rates by default and adding a Prompt Caching Dashboard, explicit cache breakpoints, and the ability to change reasoning effort or toggle tools without breaking the cache. Cached input tokens can be discounted by up to 90%.
Why it matters to your work
Practical interpretation
This is mainly relevant if your organisation builds tools or integrations on OpenAI's API — it lowers the cost of repeated or long-running AI agent calls by reusing cached context, with a new dashboard to track savings.
Source & availability
This is a summary of a published provider announcement, with practical context from SCALA’s feed.
Plan, region, workspace and rollout conditions may affect access in Australia. Check the original source for current availability.
openai.comThis update is primarily for technical teams.